本研究採用貝氏分析對加速壽命試驗模型進行分析，並將模型中參數進行重新參數化進行討論。在過去的研究中發現，針對 Type-I 設限之壽命分配，將加速壽命試驗模型中連結位置參數與應力水準函數之參數轉換為，在設限時間下正常使用應力水準與最高應力水準的累積失效機率，使其有更好的解釋性且參數範圍也更好的被定義。由於過去沒有針對加速壽命試驗之重新參數化模型進行貝氏的研究，因此本研究擬針對適當之參數化的加速壽命模型，探討設定不同的先驗分佈觀察對應之參數估計的結果。本研究針對加速壽命試驗最常使用之失效模型對數常態分佈(Lognormal) 與韋伯分佈 (Weibull) 分別進行討論，貝氏估計流程，所以採用 Gibbssampling 及 Metropolis-Hastings 方法對後驗分配做抽樣後得貝氏估計值及可信區間。先驗分佈的設定根據過去文獻設定之可能範圍去做調整，將其調整為不同程度信息的先驗分佈。結果顯示若先驗是較有信息的或參考過去經驗所得到的貝氏估計值及可信區間皆較為精確。

This study uses Bayesian analysis to analyze the accelerated life testing model and discusses the reparameterization of the model parameters. In a previous study, it is found that for the Type-I censored modeling the original parameters in the link function of the location parameter and the stress level are transformed to the cumulative failure probability of the normal use stress level and the maximum stress level at a censoring time. It resulted in a better interpretation and a better definition of the parameter ranges. Since no Bayesian analysis has been conducted for the re-parameterization of the accelerated life test in the literature, this study is intended to explore the results of the estimation of the parameters with different prior distributions of the accelerated life models with appropriate parameterization. The discussion mainly focuses on the most commonly used failure models for accelerated life testing, Lognormal distribution and Weibull distribution. Since the models have three unknown parameters, the Gibbs sampler and Metropolis-Hastings methods are used to obtain Bayesian estimates through sampling from the posterior distributions. The prior distributions are set based on the possible ranges of parameters obtained from the previous studies, by adjusting to different degrees of informative prior distributions. The results show that more accurate parameter estimates and credible intervals are obtained when the prior distributions are based on more informative prior information.

Annis, J., Miller, B. J. and Palmeri, T. J. (2017). Bayesian inference with stan: A tutorial on adding custom distributions, Behavior research methods 49: 863–886.
Ginebra, J. and Sen, A. (1998). Minimax approach to accelerated life tests, IEEE transactions on reliability 47(3): 261–267.
Gutierrez-Pulido, H., Aguirre-Torres, V. and Christen, J. (2005). A practical method for obtaining prior distributions in reliability, IEEE Transactions on Reliability 54(2): 262– 269.
Hong, Y., Ma, H. and Meeker, W. Q. (2010). A tool for evaluating time-varying-stress accelerated life test plans with log-location-scale distributions, IEEE Transactions on Reliability 59(4): 620–627.
Meeker, W. Q. (1984). A comparison of accelerated life test plans for weibull and lognormal distributions and type i censoring, Technometrics 26(2): 157–171.
Meeker, W. Q., Escobar, L. A. and Lu, C. J. (1998). Accelerated degradation tests: modeling and analysis, Technometrics 40(2): 89–99.
Nelson, W. B. (2009). Accelerated testing: statistical models, test plans, and data analysis, John Wiley & Sons.
Nelson, W. and Meeker, W. Q. (1978). Theory for optimum accelerated censored life tests for weibull and extreme value distributions, Technometrics 20(2): 171–177.
Pascual, F. G. and Montepiedra, G. (2003). Model-robust test plans with applications in accelerated life testing, Technometrics 45(1): 47–57.
Team, S. D. (2020). Rstan: the r interface to stan. r package version 2.21. 2.
Tian, Q., Lewis-Beck, C., Niemi, J. B. and Meeker, W. Q. (2023). Specifying prior dis tributions in reliability applications, Applied Stochastic Models in Business and Industry .
Weaver, B., Williams, B., Anderson-Cook, C. and Higdon, D. (2016). Computational en hancements to bayesian design of experiments using gaussian processes, Bayesian Analysis 11.
Yu, R. and Abdel-Aty, M. (2013). Investigating different approaches to develop informa tive priors in hierarchical bayesian safety performance functions, Accident Analysis & Prevention 56: 51–58.
Zhang, Y. and Meeker, W. Q. (2006). Bayesian methods for planning accelerated life tests, Technometrics 48(1): 49–60.
尹思懿 (2022). 用於搜尋加速壽命試驗中大中取小設計之混合演算法 (碩士論文)，國 立成功大學，台南市.
李馨茹 (2020). 應用粒子群最佳化方法於加速壽命試驗之大中取小設計研究 (碩士論 文)，國立成功大學，台南市.
陳凱莉 (2020). 依序貝氏程序在加速壽命試驗之初始設計規劃 (碩士論文)，國立成功 大學，台南市.